Quick prototype. I wanted to test wether an ICM7555 could be used to make a capacitive sensor by connecting an aluminum foil square as the C side of an astable configured 555 and how much of a difference in frequency would I get out of it.

Setup

Timer configured as the “alternate astable configuration” (datasheet) and the following R/C values. “Capacitor” is protected from direct contact by a piece of paper. Ballpark calculation should give me around 1nF capacitance. (It didn’t, by at least an order of magnitude, see results)

R1: 330Ω
R2: 680Ω
 C: A 225mm by 225mm paper-covered aluminium foil square

Results

Frequency/edges measurementes with a logic analyzer were as follows.

|--------------------------------+-----------+-------------+---------------|
| Description                    | Frequency | Edges @10ms | Delta vs rest |
|:-------------------------------|----------:|------------:|--------------:|
| At rest on the ground          |   1025kHz |       20516 |            0% |
| Touching with a hand           |    944kHz |       18880 |        -7.97% |
| Foot (with sock)               |    985kHz |       19714 |        -3.91% |
| Boot                           |   1014kHz |       20274 |        -1.18% |
|--------------------------------+-----------+-------------+---------------|

The difference between “at rest” and having a hand or a feet near touching it is clearly noticeable but it seems I overestimated the capacitance of the plaque and the oscillator is running around the circuit frequency limit of 1Mhz so adding an additional capacitance to move the circuit into the “linear” zone and maybe also using a total resistance equivalence of ~100k instead of 1k would yield better delta figures, making detection more reliable.

As shown on the datasheet:

Frequency against RC values

For the frequency values we had, the capacitor must lie around the 100pF value. Changing R to the 100k or 1M range would give a more spread frequency response of frequency as a function of Capacitance.